• Keine Ergebnisse gefunden

An experiment to study physiological responses to unfair pay

5.2 An experiment to study physiological

question): “In your view, how fair was the return you received from your principal?”

Answers were given on a 5-point Likert scale, with higher values indicating that returns were considered less fair.

As physiological measure of agents’ autonomic nervous system activity we used heart rate variability (HRV), an established indicator of stress-related activation of the autonomic nervous system (Task Force, 1996; Steptoe and Marmot, 2002)5. HRV reflects the continuous interaction of sympathetic and vagal influence on heart rate, indicating an individual’s capacity to generate regulated physiological responses to demanding situations (Appelhans and Luecken, 2006). Low HRV mirrors a decreased vagal tone with sympathetic predominance and is observed, among others, during states of mental stress (von Borell et al., 2007). Conversely, enhanced HRV occurs during states of mental relaxation (Vermunt and Steensma, 2003). A low HRV is an early indicator of functional and structural impairments of the cardiovascular system, which increases the probability of future manifest coronary heart disease (Steptoe and Marmot, 2002; Dekker et al., 2000; Gianaros et al., 2005). In the analysis we use two measures of HRV. The first one serves as a baseline measure (HRV_baseline) and was measured towards the end of the working period but prior to the revelation of the allocation decision. The second one was taken 15 minutes after exposure to the stimulus, i.e., the revelation of the principal’s allocation deci-sion. It records the response of the autonomic nervous system to the stimulus and

5At the beginning of the experiment a polar F810i device (polar electro OY, Kempele, Finland) was attached to record and store time intervals between consecutive heart beats (inter-beat-interval, IBI). Agents were instructed to remain seated during the whole experiment and try to restrict all movements, with the exception of their dominant arm operating the computer. The target time window for physiological recordings lasted five minutes. Data were transmitted to a PC, stored, and analyzed offline by a researcher who was blind to the psychological outcome measures. Af-ter visualizing and manually correcting data for artefacts a smoothness priors method was used to remove trends of the IBI time series. Then, a HR time series was derived and the following time-domain based HRV indices were calculated: SD-IBI (standard deviation of the IBI series), SD-HR (standard deviation of the HR series), and RMSSD-IBI (root mean square of successive differences of the IBI series) (Niskanen et al., 2004). The RMSSD-IBI represents a sensitive index of parasympathetically-dominated, respiratory related, fast fluctuations of HR, and can be calculated with milliseconds precision. It is considered to accurately index resting vagal tone directed to the heart and was documented to be rather resistant to the biasing effects of breathing (Penttilae et al., 2001). As SD-IBI and SD-HR are highly correlated with RMSSD-IBI we restrict the presentation of findings to RMSSD-IBI, as a robust and well validated time-domain based indicator of parasym-pathetic cardiac control. All calculations were done with a computer program for advanced HRV analysis (Niskanen et al., 2004).

serves as dependent variable (HRV_response).

Subjects were male students from the University of Bonn studying various ma-jors except economics. They gave their informed consent to participate in the ex-periment. Exclusion criteria were the use of medication with potential interference with cardiovascular function or the presence of a chronic disease condition, such as hypertension, cardiac arrhythmias, coronary heart disease, or diabetes. In total 80 subjects participated in the experiment (40 principals and 40 agents). During the process of data collection, we had to exclude data of 10 subjects in the role of agents, due to incomplete heart rate measurements. The main analysis is thus based on 30 subjects in the role of agents with complete data. Importantly, the 10 subjects who were excluded due to incomplete heart rate measurements were not different to the other subjects, neither in terms of working behavior nor treatment by their principals (see Footnote 7).

Experimental results. In our analysis we use three measures of perceived un-fairness, i.e., how unfair agents perceive their principals’ allocation decisions. The first measure is simply the difference between a principal’s and an agent’s payoff.

It is informed by fairness theories that model fairness comparisons in terms of de-viations from an equitable share6. Note that this measure considers wage payments and resulting payoffs only, disregarding effort costs. We have to abstract from effort costs given that in a real effort experiment, effort costs are unknown to the experi-menter. The second measure is the difference between the payoff an agent indicated as “appropriate payoff” prior to knowing the actual allocation decision, and the ac-tually received payoff. This measure therefore includes a subjective component of the agent and accounts for fairness perceptions that include both, payoffs as well as effort costs. The third measure concerns answers to the Fairness question, i.e., agents’ assessments of how fair they perceived the wage payment of their principals (on a 5-point Likert scale). This measure completely abstracts from observed wage payments and allows for a fully subjective fairness assessment of agents. It is also similar to the survey measure we use in our analysis of the effects of fairness

percep-6See, e.g., Fehr and Schmidt (1999) or Falk and Fischbacher (2006) where fairness or unfairness is evaluated as difference in payoffs (equity as a reference standard).

tions on health outcomes in the general population. The three measures are highly correlated (Spearman’s ρis between 0.498 and 0.705, p <0.01).

Table 5.1 reports means and standard deviations of our main variables7. On av-erage agents produced total revenue of 20.93 Euro and indicated that they would consider a share of 14.03 Euro (67% of total revenue) as “appropriate payoff”. This contrasts sharply with the amounts agents actually received. On average princi-pals allocated 9.53 euros to agents (46% of total revenue).8 Table 5.1 further shows the difference in payoffs of principals and agents, as well as the difference between the amounts considered as appropriate and the amounts actually received. Both differences vary considerably among subjects (standard deviations of 4.90 and 4.37, respectively). In other words the experiment generated substantial variation in (per-ceived) fairness violations, a prerequisite for the analysis of the effect of fairness perceptions on HRV.

Variable Mean Standard Deviation

Total revenue produced by agents (in Euro) 20.93 8.57

Payoff allocated to the principal (in Euro) 11.40 4.19

Payoff received by agent (in Euro) 9.53 5.58

Principal’s - agent’s payoff (in Euro) 1.87 4.90

Payoff seen as appropriate by the agent 14.03 6.68

Appropriate - actual payoff (in Euro) 4.50 4.37

Fairness question (scale: 1-5) 3.43 1.43

Table 5.1: Descriptive statistics. N = 30; appropriate refers to the amount, which is stated by the agent as appropriate pay after the total revenue was known but before the principal’s allocation decision was communicated; the difference between principal’s and agent’s payoff is our first measure of unfairness, the second is the difference between appropriate and actual payoff and the third is the answer to the Fairness question; answers are given on a 5-point Likert scale and are coded such that higher values imply higher levels of unfairness.

To test our hypothesis of an inverse relationship between the degree of fairness violation and HRV we regress HRV_response on our three measures of unfairness.

7Table 5.1 reports data for the 30 subjects with complete heart rate measurement. Subjects with incomplete measurement were not different in any systematic way. Total revenue for this group was 20.20 (Std. dev. 7.23), the payoff allocated to the principal was 11.70 (Std. dev. 3.71), the amount received by the agent 8.50 (Std. dev. 5.23) and the amount seen as appropriate by the agent was 13.80 (Std. dev. 6.34). Kruskal-Wallis rank tests do not reject the null hypothesis that both groups are drawn from the same population (p-values are between 0.54 and 0.98).

8Only two agents received more than they indicated as an appropriate amount.

The results are shown in Table 5.2. To ease comparison, the measures of unfairness are standardized. All three coefficients are negative and significant, see columns (1), (3) and (5). These results indicate that HRV reacts negatively to perceptions of being treated in an unfair way, i.e., fairness systematically affects the autonomic nervous system. Columns (2), (4) and (6) include two important control variables, HRV_baseline and generated revenue. Controlling for different baseline levels ad-dresses the possibility that subjects with a generally low baseline HRV have, e.g., systematically different fairness expectations or standards, and may therefore per-ceive payments differently. Likewise, it is important to control for levels of generated revenue to exclude the possibility that principals were willing to share relatively higher amounts with more productive agents. Results in columns (2), (4) and (6) show that our main result is robust to including these controls. While the coefficients of interest are slightly smaller compared to those reported in columns (1), (3) and (5), they remain significant.

HRV_response

(1) (2) (3) (4) (5) (6)

Principal’s - agent’s payoff -5.361∗∗ -4.717∗∗

[1.960] [1.976]

Appropriate - actual payoff -5.781∗∗∗ -4.363∗∗

[1.781] [1.773]

Fairness question -6.514∗∗∗ -5.724∗∗∗

[2.141] [1.921]

HRV_baseline 0.457∗∗∗ 0.497∗∗∗ 0.491∗∗∗

[0.145] [0.145] [0.130]

Generated revenue -0.451 -0.207 -0.369

[0.232] [0.222] [0.181]

Constant 32.072∗∗∗ 30.483∗∗∗ 32.072∗∗∗ 24.408∗∗∗ 32.072∗∗∗ 27.927∗∗∗

[1.910] [5.198] [1.868] [4.654] [1.782] [4.244]

Observations 30 30 30 30 30 30

R-squared 0.214 0.435 0.249 0.434 0.316 0.534

Table 5.2: Regression analysis on the relation between perceived fairness and HRV.

OLS estimates with robust standard errors in brackets. ∗∗∗, ∗∗, indicate sig-nificance at 1-, 5-, and 10-percent level, respectively. The dependent variable is HRV_response, i.e., the heart rate variability, which was measured after exposure to actual payoff. It records the response of the autonomic nervous system to the stimulus. HRV_baseline measures the HRV towards the end of the working period.

Appropriate refers to the amount, which is stated by the agent as appropriate pay after the total revenue was known and before the principal’s allocation decision was communicated; the difference between principal’s and agent’s payoff is our first measure of unfairness, the second is the difference between appropriate and actual payoff and the third is the answer to the Fairness question; answers are given on a 5-point Likert scale and are coded such that higher values imply higher levels of unfairness. The unfairness measures are standardized (mean = 0, standard deviation

= 1). Generated revenue represents total revenue produced by the agent.

5.3 Fairness perceptions and health: